Multi-task autoencoder based classification-regression model for patient-specific VMAT QA

Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and time-consuming. Recently, machine learning has been increasingly investigated in PSQA results prediction. However, the classification performance of mo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Physics in medicine & biology Jg. 65; H. 23; S. 235023
Hauptverfasser: Wang, Le, Li, Jiaqi, Zhang, Shuming, Zhang, Xile, Zhang, Qilin, Chan, Maria F, Yang, Ruijie, Sui, Jing
Format: Journal Article
Sprache:Englisch
Veröffentlicht: England 27.11.2020
Schlagworte:
ISSN:1361-6560, 1361-6560
Online-Zugang:Weitere Angaben
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and time-consuming. Recently, machine learning has been increasingly investigated in PSQA results prediction. However, the classification performance of models at different criteria needs further improvement and clinical validation (CV), especially for predicting plans with low gamma passing rates (GPRs). In this study, we developed and validated a novel multi-task model called autoencoder based classification-regression (ACLR) for VMAT PSQA. The classification and regression were integrated into one model, both parts were trained alternatively while minimizing a defined loss function. The classification was used as an intermediate result to improve the regression accuracy. Different tasks of GPRs prediction and classification based on different criteria were trained simultaneously. Balanced sampling techniques were used to improve the prediction accuracy and classification sensitivity for the unbalanced VMAT plans. Fifty-four metrics were selected as inputs to describe the plan modulation-complexity and delivery-characteristics, while the outputs were PSQA GPRs. A total of 426 clinically delivered VMAT plans were used for technical validation (TV), and another 150 VMAT plans were used for CV to evaluate the generalization performance of the model. The ACLR performance was compared with the Poisson Lasso (PL) model and found significant improvement in prediction accuracy. In TV, the absolute prediction error (APE) of ACLR was 1.76%, 2.60%, and 4.66% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively; whereas the APE of PL was 2.10%, 3.04%, and 5.29% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. No significant difference was found between CV and TV in prediction accuracy. ACLR model set with 3%/3 mm can achieve 100% sensitivity and 83% specificity. The ACLR model could classify the unbalanced VMAT QA results accurately, and it can be readily applied in clinical practice for virtual VMAT QA.
AbstractList Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and time-consuming. Recently, machine learning has been increasingly investigated in PSQA results prediction. However, the classification performance of models at different criteria needs further improvement and clinical validation (CV), especially for predicting plans with low gamma passing rates (GPRs). In this study, we developed and validated a novel multi-task model called autoencoder based classification-regression (ACLR) for VMAT PSQA. The classification and regression were integrated into one model, both parts were trained alternatively while minimizing a defined loss function. The classification was used as an intermediate result to improve the regression accuracy. Different tasks of GPRs prediction and classification based on different criteria were trained simultaneously. Balanced sampling techniques were used to improve the prediction accuracy and classification sensitivity for the unbalanced VMAT plans. Fifty-four metrics were selected as inputs to describe the plan modulation-complexity and delivery-characteristics, while the outputs were PSQA GPRs. A total of 426 clinically delivered VMAT plans were used for technical validation (TV), and another 150 VMAT plans were used for CV to evaluate the generalization performance of the model. The ACLR performance was compared with the Poisson Lasso (PL) model and found significant improvement in prediction accuracy. In TV, the absolute prediction error (APE) of ACLR was 1.76%, 2.60%, and 4.66% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively; whereas the APE of PL was 2.10%, 3.04%, and 5.29% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. No significant difference was found between CV and TV in prediction accuracy. ACLR model set with 3%/3 mm can achieve 100% sensitivity and 83% specificity. The ACLR model could classify the unbalanced VMAT QA results accurately, and it can be readily applied in clinical practice for virtual VMAT QA.
Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and time-consuming. Recently, machine learning has been increasingly investigated in PSQA results prediction. However, the classification performance of models at different criteria needs further improvement and clinical validation (CV), especially for predicting plans with low gamma passing rates (GPRs). In this study, we developed and validated a novel multi-task model called autoencoder based classification-regression (ACLR) for VMAT PSQA. The classification and regression were integrated into one model, both parts were trained alternatively while minimizing a defined loss function. The classification was used as an intermediate result to improve the regression accuracy. Different tasks of GPRs prediction and classification based on different criteria were trained simultaneously. Balanced sampling techniques were used to improve the prediction accuracy and classification sensitivity for the unbalanced VMAT plans. Fifty-four metrics were selected as inputs to describe the plan modulation-complexity and delivery-characteristics, while the outputs were PSQA GPRs. A total of 426 clinically delivered VMAT plans were used for technical validation (TV), and another 150 VMAT plans were used for CV to evaluate the generalization performance of the model. The ACLR performance was compared with the Poisson Lasso (PL) model and found significant improvement in prediction accuracy. In TV, the absolute prediction error (APE) of ACLR was 1.76%, 2.60%, and 4.66% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively; whereas the APE of PL was 2.10%, 3.04%, and 5.29% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. No significant difference was found between CV and TV in prediction accuracy. ACLR model set with 3%/3 mm can achieve 100% sensitivity and 83% specificity. The ACLR model could classify the unbalanced VMAT QA results accurately, and it can be readily applied in clinical practice for virtual VMAT QA.Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and time-consuming. Recently, machine learning has been increasingly investigated in PSQA results prediction. However, the classification performance of models at different criteria needs further improvement and clinical validation (CV), especially for predicting plans with low gamma passing rates (GPRs). In this study, we developed and validated a novel multi-task model called autoencoder based classification-regression (ACLR) for VMAT PSQA. The classification and regression were integrated into one model, both parts were trained alternatively while minimizing a defined loss function. The classification was used as an intermediate result to improve the regression accuracy. Different tasks of GPRs prediction and classification based on different criteria were trained simultaneously. Balanced sampling techniques were used to improve the prediction accuracy and classification sensitivity for the unbalanced VMAT plans. Fifty-four metrics were selected as inputs to describe the plan modulation-complexity and delivery-characteristics, while the outputs were PSQA GPRs. A total of 426 clinically delivered VMAT plans were used for technical validation (TV), and another 150 VMAT plans were used for CV to evaluate the generalization performance of the model. The ACLR performance was compared with the Poisson Lasso (PL) model and found significant improvement in prediction accuracy. In TV, the absolute prediction error (APE) of ACLR was 1.76%, 2.60%, and 4.66% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively; whereas the APE of PL was 2.10%, 3.04%, and 5.29% at 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. No significant difference was found between CV and TV in prediction accuracy. ACLR model set with 3%/3 mm can achieve 100% sensitivity and 83% specificity. The ACLR model could classify the unbalanced VMAT QA results accurately, and it can be readily applied in clinical practice for virtual VMAT QA.
Author Sui, Jing
Li, Jiaqi
Chan, Maria F
Zhang, Xile
Wang, Le
Yang, Ruijie
Zhang, Shuming
Zhang, Qilin
Author_xml – sequence: 1
  givenname: Le
  surname: Wang
  fullname: Wang, Le
  organization: Contributed equally to this work
– sequence: 2
  givenname: Jiaqi
  surname: Li
  fullname: Li, Jiaqi
  organization: Contributed equally to this work
– sequence: 3
  givenname: Shuming
  surname: Zhang
  fullname: Zhang, Shuming
  organization: Department of Radiation Oncology, Peking University Third Hospital, Beijing, People's Republic of China
– sequence: 4
  givenname: Xile
  surname: Zhang
  fullname: Zhang, Xile
  organization: Department of Radiation Oncology, Peking University Third Hospital, Beijing, People's Republic of China
– sequence: 5
  givenname: Qilin
  surname: Zhang
  fullname: Zhang, Qilin
  organization: Department of Radiation Oncology, Peking University Third Hospital, Beijing, People's Republic of China
– sequence: 6
  givenname: Maria F
  surname: Chan
  fullname: Chan, Maria F
  organization: Medical Physics Department, Memorial Sloan Kettering Cancer Center, New York, NY, United States of America
– sequence: 7
  givenname: Ruijie
  orcidid: 0000-0002-3459-0075
  surname: Yang
  fullname: Yang, Ruijie
  organization: Author to whom any correspondence should be addressed
– sequence: 8
  givenname: Jing
  orcidid: 0000-0001-6837-5966
  surname: Sui
  fullname: Sui, Jing
  organization: Author to whom any correspondence should be addressed
BackLink https://www.ncbi.nlm.nih.gov/pubmed/33245054$$D View this record in MEDLINE/PubMed
BookMark eNpNkMtLxDAYxIOsuA-9e5IcvdRN0iTNHpfFF-wiwip4Knl8kWrb1KQ9-N9bcRVPM8z8mMPM0aQNLSB0TskVJUotaS5pJoUkS21MTu0Rmv1Fk39-iuYpvRFCqWL8BE3znHFBBJ-hl91Q91XW6_SO9dAHaG1wELHRCRy2tU6p8pXVfRXaLMJrhDEILW5GqsY-RNyNHbR9ljqw3yh-3q33-HF9io69rhOcHXSBnm6u95u7bPtwe79ZbzMrVrTPqOSFE4XOpQanlDdEeybBGGKNgQKI16wwK6Gd8gq4sysuZa6VZ05YYghboMuf3S6GjwFSXzZVslDXuoUwpJJxKTgnlMsRvTigg2nAlV2sGh0_y9872BeoFmWJ
CitedBy_id crossref_primary_10_1016_j_ejmp_2023_102593
crossref_primary_10_1186_s13014_022_01999_3
crossref_primary_10_1002_mp_17485
crossref_primary_10_1002_mp_16210
crossref_primary_10_1016_j_ejmp_2024_103208
crossref_primary_10_1016_j_phro_2023_100525
crossref_primary_10_3390_bioengineering12080897
crossref_primary_10_1088_2057_1976_acd255
crossref_primary_10_1186_s13014_023_02376_4
crossref_primary_10_1088_1361_6560_adf40b
crossref_primary_10_3389_fonc_2021_700343
crossref_primary_10_1016_j_apradiso_2025_112030
crossref_primary_10_1016_j_radonc_2021_06_024
crossref_primary_10_1002_mp_17081
crossref_primary_10_3390_diagnostics13050943
crossref_primary_10_1016_j_clon_2021_11_011
crossref_primary_10_1016_j_ejmp_2024_104500
crossref_primary_10_1016_j_ejmp_2021_05_007
crossref_primary_10_1002_mp_15031
crossref_primary_10_1002_mp_17298
crossref_primary_10_1002_mp_15393
crossref_primary_10_1016_j_radonc_2022_05_016
crossref_primary_10_1016_j_radonc_2025_110934
crossref_primary_10_1016_j_ijrobp_2022_04_040
crossref_primary_10_3233_XST_230412
crossref_primary_10_1002_acm2_13375
crossref_primary_10_3389_fdata_2024_1462745
crossref_primary_10_1016_j_anucene_2024_110444
crossref_primary_10_1155_2022_5448442
crossref_primary_10_3389_fonc_2023_1142947
ContentType Journal Article
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1088/1361-6560/abb31c
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod no_fulltext_linktorsrc
Discipline Medicine
Biology
Physics
EISSN 1361-6560
ExternalDocumentID 33245054
Genre Research Support, Non-U.S. Gov't
Journal Article
Research Support, N.I.H., Extramural
GrantInformation_xml – fundername: NCI NIH HHS
  grantid: P30 CA008748
GroupedDBID ---
-DZ
-~X
123
1JI
4.4
5B3
5RE
5VS
5ZH
7.M
7.Q
AAGCD
AAJIO
AAJKP
AATNI
ABCXL
ABHWH
ABJNI
ABLJU
ABQJV
ABVAM
ACAFW
ACGFS
ACHIP
AEFHF
AENEX
AFYNE
AKPSB
ALMA_UNASSIGNED_HOLDINGS
AOAED
ASPBG
ATQHT
AVWKF
AZFZN
CBCFC
CEBXE
CGR
CJUJL
CRLBU
CS3
CUY
CVF
DU5
EBS
ECM
EDWGO
EIF
EJD
EMSAF
EPQRW
EQZZN
F5P
HAK
IHE
IJHAN
IOP
IZVLO
KOT
LAP
M45
N5L
N9A
NPM
P2P
PJBAE
R4D
RIN
RNS
RO9
ROL
RPA
SY9
TN5
UCJ
W28
XPP
7X8
ADEQX
AEINN
ID FETCH-LOGICAL-c591t-1647d57a36aed88fb0af26ebb0cbbe7e0fa27b95ad8f8e4dc94663a8f2d5c0b02
IEDL.DBID 7X8
ISICitedReferencesCount 37
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000592683300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1361-6560
IngestDate Fri Sep 05 13:37:07 EDT 2025
Thu Jan 02 22:53:43 EST 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 23
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c591t-1647d57a36aed88fb0af26ebb0cbbe7e0fa27b95ad8f8e4dc94663a8f2d5c0b02
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-3459-0075
0000-0001-6837-5966
OpenAccessLink https://pmc.ncbi.nlm.nih.gov/articles/PMC10072931/pdf/nihms-1883265.pdf
PMID 33245054
PQID 2465440146
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2465440146
pubmed_primary_33245054
PublicationCentury 2000
PublicationDate 20201127
PublicationDateYYYYMMDD 2020-11-27
PublicationDate_xml – month: 11
  year: 2020
  text: 20201127
  day: 27
PublicationDecade 2020
PublicationPlace England
PublicationPlace_xml – name: England
PublicationTitle Physics in medicine & biology
PublicationTitleAlternate Phys Med Biol
PublicationYear 2020
SSID ssj0011824
Score 2.4967268
Snippet Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) to assure accurate treatment delivery is resource-intensive and...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 235023
SubjectTerms Humans
Machine Learning
Quality Control
Radiotherapy Planning, Computer-Assisted - methods
Radiotherapy, Intensity-Modulated - methods
Title Multi-task autoencoder based classification-regression model for patient-specific VMAT QA
URI https://www.ncbi.nlm.nih.gov/pubmed/33245054
https://www.proquest.com/docview/2465440146
Volume 65
WOSCitedRecordID wos000592683300001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LSwMxEA5qVbz4qK_6IoLX0E32kexJili8tFSoUk8lrxURdmt3K_jvzeym9iQIXvYQNhCSycyXzOT7ELoJFOUZZYoYziICGtbODxpBuGGBhBCTeLEJPhyKySQd-Qu30pdVLn1i7ahNoeGOvMuA-CsCqpPb2QcB1SjIrnoJjXXUCh2UgZIuPlllERx2bkRtE0qAZManKd3G6v60gWZXSPXvALMONP29_w5xH-16iIl7jU0coDWbt9FWIzr51UbbA59Od411_acuD9FL_RCXVLJ8x3JRFUBvaewcQ5AzWAPEhpqiehnJ3L421bM5roV0sAO-2BO0Eni6Cb_i50FvjB97R-ipfz--eyBedYHoOKUVAYIxE3MZJtIaITIVyIwlVqlAK2W5DTLJuEpjaUQmbGQ0MNSHUmTMxDpQATtGG3mR21OEkziNQqMSbRyKoDSV2uEhMBqhueTcdtD1ciKnzqohVSFzWyzK6WoqO-ikWY3prKHfmIYOAzrcFp39ofc52mFwQKaUMH6BWpnb0_YSberP6q2cX9Xm4r7D0eAbAGzJuQ
linkProvider ProQuest
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Multi-task+autoencoder+based+classification-regression+model+for+patient-specific+VMAT+QA&rft.jtitle=Physics+in+medicine+%26+biology&rft.au=Wang%2C+Le&rft.au=Li%2C+Jiaqi&rft.au=Zhang%2C+Shuming&rft.au=Zhang%2C+Xile&rft.date=2020-11-27&rft.issn=1361-6560&rft.eissn=1361-6560&rft.volume=65&rft.issue=23&rft.spage=235023&rft_id=info:doi/10.1088%2F1361-6560%2Fabb31c&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1361-6560&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1361-6560&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1361-6560&client=summon